Implementation brief

## Who Uses It
An on-call engineer investigating service degradation.

## What You Will Build
Build an agent that combines alerts, metrics, logs, and deployment history into an incident timeline. Recommend a recovery plan, request approval, and verify recovery in a local service simulator.

## Inputs
- Alert feed
- Metrics and log records
- Deployment history and runbook

## Workflow
1. Group related alerts into an incident.
2. Build a timeline and compare plausible causes.
3. Present a runbook action for approval.
4. Execute the approved action in the simulator and check recovery.

## Interface
An incident console with evidence links, an approval button, recovery charts, and a downloadable incident report.

## Working Demo
Explain the deployment-linked incident, request rollback approval, simulate recovery, and produce a cited timeline.

## Deliverables
- A working local application with the specified review interface.
- Source code, dependency lockfile, and setup commands.
- A recorded end-to-end demo using the supplied starter data.
- Example outputs, action logs, and automated tests.

## Starter Data
These fictional records are a small development fixture, not a production dataset. Implement the mock services described in the brief.

```json
{
  "request": "Investigate checkout latency.",
  "deployments": [
    {
      "id": "D1",
      "service": "checkout",
      "minute": 5
    }
  ],
  "metrics": [
    {
      "minute": 4,
      "latency_ms": 120
    },
    {
      "minute": 6,
      "latency_ms": 900
    }
  ],
  "logs": [
    {
      "id": "L1",
      "minute": 6,
      "text": "New checkout release exhausts its connection pool."
    }
  ],
  "runbook": {
    "action": "rollback_D1",
    "requires_approval": true,
    "recovery_limit_ms": 200
  },
  "simulator_after_rollback_ms": [
    150,
    140,
    130
  ]
}
```

## Constraints
Use mock service controls. Do not connect production cluster credentials.
Choose an available model that supports your implementation. Record its identifier and version.

## Evaluation
Review the complete application, not only the starter-case answers.
- The timeline references D1 and L1.
- No rollback occurs before approval.
- Recovery is reported only after all three supplied recovery readings pass.
Add one normal case, one incomplete-input case, and one failure-recovery case.